Face SDK v3.8.0

3DiVi · Feb 5, 2021 · 94b8901 · 94b8901
1 parent 841ea67
commit 94b8901
Show file tree

Hide file tree

Showing 124 changed files with 1,501 additions and 201 deletions.
diff --git a/.gitattributes b/.gitattributes
@@ -9,4 +9,4 @@ linux_x86_32.tar.xz filter=lfs diff=lfs merge=lfs -text
 windows_x86_32.zip filter=lfs diff=lfs merge=lfs -text
 linux_astra_1.11_x86_64.tar.xz filter=lfs diff=lfs merge=lfs -text
 share/** filter=lfs diff=lfs merge=lfs -text
-linux_x86_64_cuda.tar.xz filter=lfs diff=lfs merge=lfs -text
+jetson_jetpack_4.3_4.4.tar.xz filter=lfs diff=lfs merge=lfs -text
diff --git a/README.md b/README.md
@@ -34,7 +34,7 @@ Face SDK provides the following APIs:
 
   * [Components](doc/en/components.md) – essential information about the components included in Face SDK
   * [Use Cases](doc/en/use_cases.md) – diagrams that show possible use cases and connection between the components
-  * [Samples](doc/en/samples) – sample programs in C++/C#/Java/Python/Javascript that show you the features of Face SDK
+  * [Samples](doc/en/samples) – sample programs in C++/C#/Java/Python/Javascript and a Unity sample that show you the features of Face SDK
   * [Tutorials](doc/en/tutorials) – step-by-step tutorials on face detection, recognition, and estimation of age, gender, and emotions
   * [Licenses](doc/en/licenses.md) – information about different types of licenses, activation and locking of a license
   * [Performance Parameters](doc/en/performance_parameters.md) – CPU and memory usage, identification quality (FAR/FRR)
@@ -86,7 +86,8 @@ When extracting the archive, specify the path to the Face SDK root – the folde
 * [Error Handling](doc/en/development/error_handling.md) – error handling in C++/C#/Java/Python
 * [Memory Management](doc/en/development/memory_management.md) – memory management in C++/C#/Java/Python
 * [Guidelines for Cameras](doc/en/guidelines_for_cameras.md) – camera positioning and shooting, recommended cameras
-* [Face SDK Cross-Platform API. Latest Doxygen Output](http://download.3divi.com/facesdk/0d88ba7c-9a5d-45cd-897a-406fb1fca2d4/latest_docs/english/annotated.html) 
+* [Face SDK Cross-Platform API. Latest Doxygen Output](http://download.3divi.com/facesdk/archives/latest/docs/english/annotated.html) 
+* [Troubleshooting](/doc/en/development/troubleshooting.md) – error codes and description, possible solutions 
 
 # How can I propose changes to Face SDK? 
 

diff --git a/README_rus.md b/README_rus.md
@@ -34,7 +34,7 @@ Face SDK включает в себя следующие API:
 
 * [Компоненты](doc/ru/components.md) – важная информация о компонентах Face SDK
 * [Сценарии использования](/doc/ru/use_cases.md) – схемы, показывающие возможные сценарии использования и связь между компонентами
-* [Сэмплы](doc/ru/samples) – программы на языках C++/C#/Java/Python/Javascript, демонстрирующие функции Face SDK
+* [Сэмплы](doc/ru/samples) – программы на языках C++/C#/Java/Python/Javascript и Unity-сэмпл, демонстрирующие функции Face SDK
 * [Туториалы](doc/ru/tutorials) – пошаговые туториалы по разработке проектов с трекингом и распознаванием лиц, а также определением пола, возраста и эмоций лица
 * [Лицензии](doc/ru/licenses.md) – информация о разных типах лицензий, активации и привязке лицензий 
 * [Параметры производительности](doc/ru/performance_parameters.md) – загрузка ЦП и использование памяти, качество идентификации (FAR/FRR) 
@@ -84,7 +84,8 @@ $ git lfs pull
 * [Обработка ошибок](/doc/ru/development/error_handling.md) – обработка ошибок в C++/C#/Java/Python
 * [Управление памятью](/doc/ru/development/memory_management.md) – управление памятью в C++/C#/Java/Python 
 * [Рекомендации по камерам](/doc/ru/guidelines_for_cameras.md) – размещение камер и съемка, рекомендуемые характеристики камер
-* [Межплатформенное Face SDK API. Последняя версия в формате Doxygen](http://download.3divi.com/facesdk/0d88ba7c-9a5d-45cd-897a-406fb1fca2d4/latest_docs/russian/annotated.html)
+* [Межплатформенное Face SDK API. Последняя версия в формате Doxygen](http://download.3divi.com/facesdk/archives/latest/docs/russian/annotated.html)
+* [Диагностика и устранение неполадок](/doc/ru/development/troubleshooting.md) – коды ошибок, сообщения об ошибках и возможные решения
 
 # Как я могу предложить изменения в Face SDK? 
 

diff --git a/RELEASE_NOTES.md b/RELEASE_NOTES.md
@@ -1,32 +1,26 @@
 # Release Notes
 
-## 3.7.1 (18 Jan 2021)
-
-### Bug Fixes and Improvements
-
-* Fixed the reading error from 3divi_face_sdk.lic when using a USB token
-
-
-## 3.6.1 (18 Jan 2021)
-
-### Bug Fixes and Improvements
-
-* Fixed fake detections when processing video from multiple streams
-* Fixed the reading error from 3divi_face_sdk.lic when using a USB token
+## 3.8.0 (28 Jan 2021) 
 
+### What's new
 
-## 3.5.1 (18 Jan 2021)
+* Added display of the used FaceSDK library version
+* Added GPU support on Android for liveness_2d_estimator.xml
+* Added GPU support on Jetson (JetPack 4.3/4.4)
+* Added the FaceMask Unity sample on Windows and Android
+* Added the possibility to estimate the presence of a mask on the face
+* Added a new version of RGB Passive Liveness (liveness_2d_estimator_v2.xml)
 
 ### Bug Fixes and Improvements
 
-* Fixed fake detections when processing video from multiple streams
+* Added display of the score parameter for Liveness2DEstimator
+* Added base_angle support for the ULD, BLD, REFA detectors in FDATracker
 
-
-## 3.4.2 (18 Jan 2021)
+## 3.7.1 (18 Jan 2021)
 
 ### Bug Fixes and Improvements
 
-* Fixed fake detections when processing video from multiple streams
+* Fixed the reading error from 3divi_face_sdk.lic when using a USB token
 
 
 ## v3.7.0 (21 Dec 2020)

diff --git a/android_arm_32.tar.xz b/android_arm_32.tar.xz
diff --git a/android_arm_64.tar.xz b/android_arm_64.tar.xz
diff --git a/conf/facerec/common_capturer_blf_fda_back.xml b/conf/facerec/common_capturer_blf_fda_back.xml
@@ -15,7 +15,10 @@
 
 <blf_id>blf-back</blf_id>
 <blf_detector_file>share/facedetectors/blf/back_model.enc</blf_detector_file>
+
 <use_mobile_gpu>1</use_mobile_gpu>
+<use_cuda>0</use_cuda>
+<gpu_index>0</gpu_index>
 
 
 <fitter_type>fda</fitter_type>

diff --git a/conf/facerec/common_capturer_blf_fda_front.xml b/conf/facerec/common_capturer_blf_fda_front.xml
@@ -15,7 +15,10 @@
 
 <blf_id>blf-front</blf_id>
 <blf_detector_file>share/facedetectors/blf/front_model.enc</blf_detector_file>
+
 <use_mobile_gpu>0</use_mobile_gpu>
+<use_cuda>0</use_cuda>
+<gpu_index>0</gpu_index>
 
 
 <fitter_type>fda</fitter_type>

diff --git a/conf/facerec/common_capturer_refa_fda_a.xml b/conf/facerec/common_capturer_refa_fda_a.xml
@@ -17,6 +17,8 @@
 <refa_id>refa-a</refa_id>
 <refa_filepath>share/refa/refa-a.enc</refa_filepath>
 
+<use_cuda>0</use_cuda>
+<gpu_index>0</gpu_index>
 
 <fitter_type>fda</fitter_type>
 <fda_file>share/fda/fda0017.3500.enc</fda_file>

diff --git a/conf/facerec/common_capturer_uld_fda.xml b/conf/facerec/common_capturer_uld_fda.xml
@@ -21,6 +21,9 @@
 <score_threshold>0.7</score_threshold>
 <mask_score_threshold>1.0</mask_score_threshold>
 
+<use_cuda>0</use_cuda>
+<gpu_index>0</gpu_index>
+
 <fitter_type>fda</fitter_type>
 <fda_file>share/fda/fda0017.3500.enc</fda_file>
 

diff --git a/conf/facerec/face_mask_estimator.xml b/conf/facerec/face_mask_estimator.xml
@@ -0,0 +1,12 @@
+<?xml version="1.0"?>
+<opencv_storage>
+
+<face_attribute_impl>the_one</face_attribute_impl>
+
+<face_attribute_name>masked_face</face_attribute_name>
+<face_attribute_data>share/faceattributes/face_mask.enc</face_attribute_data>
+<score_threshold>0.0001498</score_threshold>
+
+<use_mobile_gpu>0</use_mobile_gpu>
+
+</opencv_storage>
diff --git a/conf/facerec/fda_tracker_capturer_blf.xml b/conf/facerec/fda_tracker_capturer_blf.xml
@@ -13,6 +13,9 @@
 <blf_id>blf-back</blf_id>
 <blf_detector_file>share/facedetectors/blf/back_model.enc</blf_detector_file>
 
+<use_cuda>0</use_cuda>
+<gpu_index>0</gpu_index>
+
 
 <fda_max_bad_count_wait>5</fda_max_bad_count_wait>
 <fda_max_weak_count_wait>30</fda_max_weak_count_wait>

diff --git a/conf/facerec/fda_tracker_capturer_blf_front.xml b/conf/facerec/fda_tracker_capturer_blf_front.xml
@@ -13,6 +13,9 @@
 <blf_id>blf-back</blf_id>
 <blf_detector_file>share/facedetectors/blf/back_model.enc</blf_detector_file>
 
+<use_cuda>0</use_cuda>
+<gpu_index>0</gpu_index>
+
 
 <fda_max_bad_count_wait>5</fda_max_bad_count_wait>
 <fda_max_weak_count_wait>30</fda_max_weak_count_wait>

diff --git a/conf/facerec/fda_tracker_capturer_refa_a.xml b/conf/facerec/fda_tracker_capturer_refa_a.xml
@@ -15,6 +15,9 @@
 <refa_id>refa-a</refa_id>
 <refa_filepath>share/refa/refa-a.enc</refa_filepath>
 
+<use_cuda>0</use_cuda>
+<gpu_index>0</gpu_index>
+
 
 <fda_max_bad_count_wait>5</fda_max_bad_count_wait>
 <fda_max_weak_count_wait>30</fda_max_weak_count_wait>

diff --git a/conf/facerec/fda_tracker_capturer_uld.xml b/conf/facerec/fda_tracker_capturer_uld.xml
@@ -22,6 +22,9 @@
 <score_threshold>0.4</score_threshold>
 <mask_score_threshold>1.0</mask_score_threshold>
 
+<use_cuda>0</use_cuda>
+<gpu_index>0</gpu_index>
+
 
 <fda_max_bad_count_wait>5</fda_max_bad_count_wait>
 <fda_max_weak_count_wait>30</fda_max_weak_count_wait>

diff --git a/conf/facerec/liveness_2d_estimator.xml b/conf/facerec/liveness_2d_estimator.xml
@@ -6,6 +6,6 @@
 <liveness_2d_impl>l-0018</liveness_2d_impl>
 <liveness_2d_iou_threshold>0.5</liveness_2d_iou_threshold>
 <liveness_2d_score_threshold>0.5</liveness_2d_score_threshold>
-<liveness_2d_num_threads>4</liveness_2d_num_threads>
+<num_threads>4</num_threads>
 
 </opencv_storage>
diff --git a/conf/facerec/liveness_2d_estimator_v2.xml b/conf/facerec/liveness_2d_estimator_v2.xml
@@ -0,0 +1,9 @@
+<?xml version="1.0"?>
+<opencv_storage>
+
+<liveness_2d_impl_name>the_two</liveness_2d_impl_name>
+<liveness_2d_data>share/liveness_2d_v2/model.enc</liveness_2d_data>
+<liveness_2d_impl>liveness_2d_v2</liveness_2d_impl>
+<liveness_2d_score_threshold>0.9</liveness_2d_score_threshold>
+
+</opencv_storage>
diff --git a/conf/facerec/video_worker_fdatracker_blf_fda.xml b/conf/facerec/video_worker_fdatracker_blf_fda.xml
@@ -36,6 +36,9 @@
 <blf_detector_file>share/facedetectors/blf/back_model.enc</blf_detector_file>
 <use_mobile_gpu>1</use_mobile_gpu>
 
+<use_cuda>0</use_cuda>
+<gpu_index>0</gpu_index>
+
 
 <depth_data_flag>0</depth_data_flag>
 <depth_liveness_config>conf/facerec/depth_liveness_estimator_cnn.xml</depth_liveness_config>

diff --git a/conf/facerec/video_worker_fdatracker_blf_fda_front.xml b/conf/facerec/video_worker_fdatracker_blf_fda_front.xml
@@ -36,6 +36,9 @@
 <blf_detector_file>share/facedetectors/blf/front_model.enc</blf_detector_file>
 <use_mobile_gpu>0</use_mobile_gpu>
 
+<use_cuda>0</use_cuda>
+<gpu_index>0</gpu_index>
+
 
 <depth_data_flag>0</depth_data_flag>
 <depth_liveness_config>conf/facerec/depth_liveness_estimator_cnn.xml</depth_liveness_config>

diff --git a/conf/facerec/video_worker_fdatracker_refa_fda.xml b/conf/facerec/video_worker_fdatracker_refa_fda.xml
@@ -34,6 +34,9 @@
 <refa_id>refa-a</refa_id>
 <refa_filepath>share/refa/refa-a.enc</refa_filepath>
 
+<use_cuda>0</use_cuda>
+<gpu_index>0</gpu_index>
+
 
 <depth_data_flag>0</depth_data_flag>
 <depth_liveness_config>conf/facerec/depth_liveness_estimator_cnn.xml</depth_liveness_config>

diff --git a/conf/facerec/video_worker_fdatracker_uld_fda.xml b/conf/facerec/video_worker_fdatracker_uld_fda.xml
@@ -39,6 +39,9 @@
 <score_threshold>0.7</score_threshold>
 <mask_score_threshold>1.0</mask_score_threshold>
 
+<use_cuda>0</use_cuda>
+<gpu_index>0</gpu_index>
+
 
 <depth_data_flag>0</depth_data_flag>
 <depth_liveness_config>conf/facerec/depth_liveness_estimator_cnn.xml</depth_liveness_config>

diff --git a/doc/en/development/connect_facesdk.md b/doc/en/development/connect_facesdk.md
@@ -29,10 +29,12 @@ The *libfacerec* library does not have any actual third-party dependencies*, sin
 
 #### Library Version
 
-Following preprocessor constants specify the current library version:
+You can get the current library version by calling the `get_version` method from the created `FacerecService` object. This will return a string in the format of `x.yy.zz`, where `x` is the major version, `yy` is the minor version, and `zz` is the patch version.
 
-* `LIBFACEREC_VERSION` – string representation, e.g. "1.2.34", where 1 is the major version, 2 is the minor version, 34 is the patch version
-* `LIBFACEREC_VERSION_HEX` – 3-byte HEX representation that is useful in numeric comparisons, e.g. 0x010234
+In C++, the version can be additionally obtained from the preprocessor constants:
+
+* `LIBFACEREC_VERSION` – string representation, similar to `FacerecService.get_version()`
+* `LIBFACEREC_VERSION_HEX` – 3-byte HEX representation that is useful in numeric comparisons, e.g. `0x010234`
 
 ### Java
 

diff --git a/doc/en/development/face_capturing.md b/doc/en/development/face_capturing.md
@@ -292,7 +292,9 @@ capturer.setParameter("max_size", 400);
 |manual_capturer_fda.xml|lbf|fda|[-30;30][-60;60][-60;60]|Eye points should be manually specified. The remaining points are calculated based on the eye points.|
 |manual_capturer_mesh.xml|lbf|mesh|[-30;30][-60;60][-60;60]|Eye points should be manually specified. The remaining points are calculated based on the eye points. Allows you to get a 3D face mask.|
 
-### Capturers Timing Characteristics for Core i7 4.5 GHz*
+### Capturers Timing Characteristics 
+
+#### Core i7 4.5 GHz (Single-Core)
 
 <table>
 <thead>
@@ -364,7 +366,53 @@ capturer.setParameter("max_size", 400);
   </tr>
 </table>
 
-\* – characteristics specified in this table are given for a single-core CPU 
+#### GPU 
+
+<table>
+<thead>
+  <tr>
+    <th rowspan="2">config file</th>
+    <th colspan="6">capture time (ms)</th>
+  </tr>
+  <tr>
+    <td>640x480, 1 face</td>
+    <td>640x480, 4 faces</td>
+    <td>1280x720, 1 face</td>
+    <td>1280x720, 4 faces</td>
+    <td>1920x1080, 1 face</td>
+    <td>1920x1080, 4 faces</td>
+  </tr>
+</thead>
+<tbody>
+  <tr>
+    <td>common_capturer_blf_fda_auto.xml</td>
+    <td>4-5</td> <td>10-12</td> <td>6-8</td> <td>13-14</td> <td>17-20</td> <td>24-27</td>
+  </tr>
+  <tr>
+    <td>common_capturer_blf_fda_back.xml</td>
+    <td>5</td> <td>12</td> <td>8</td> <td>14</td> <td>20</td> <td>27</td>
+  </tr>
+  <tr>
+    <td>common_capturer_blf_fda_front.xml</td>
+    <td>4</td> <td>10</td> <td>6</td> <td>13</td> <td>17</td> <td>24</td>
+  </tr>
+  <tr>
+    <td>common_capturer_refa_fda_a.xml</td>
+    <td>236</td> <td>240</td> <td>229</td> <td>235</td> <td>170</td> <td>176</td>
+  </tr>
+  <tr>
+    <td>common_capturer_uld_fda.xml (min_size=150)</td>
+    <td>4</td> <td>10</td> <td>5</td> <td>11</td> <td>13</td> <td>20</td>
+  </tr>
+  <tr>
+    <td>common_capturer_uld_fda.xml (min_size=90)</td>
+    <td>14</td> <td>21</td> <td>17</td> <td>23</td> <td>26</td> <td>34</td>
+  </tr>
+  <tr>
+    <td>common_capturer_uld_fda.xml (min_size=50)</td>
+    <td>27</td> <td>34</td> <td>27</td> <td>35</td> <td>47</td> <td>49</td>
+  </tr>
+</table>
 
 _**Note:** Actual capture time may vary depending on the image content._